NHS hospital care: Who is waiting and what are they waiting for?

Backlogs in NHS care and long waiting times in England are widespread, and politicians, policymakers and the public are well aware of this. But headline numbers obscure important distinctions, and questions remain unanswered about who is bearing the brunt of delays. This new QualityWatch analysis uses urgent and emergency care and planned care data alongside an ONS patient survey to understand how waiting times vary by age, sex, ethnicity, and level of deprivation, and how waits for specific conditions differ.

Qualitywatch

Long read

Published: 10/10/2024

Long waiting times are a persistent problem for the health service in England, provoking public dissatisfaction and political scrutiny. Reducing these waits is a top priority for the new government, with pledges for new appointments forming a key part of their recent election manifesto.

But much of the commentary on NHS waits – including our own monthly NHS performance tracker – focuses on the overall numbers of people waiting for care and waiting times across all those waiting. These headline figures hide variations and leave important questions unanswered about who is bearing the brunt of delays.

This analysis takes a closer look at NHS waits, delving into data 1  on urgent and emergency care and planned care segmented by what people are waiting for and who is waiting by age, sex, ethnicity, and relative deprivation levels. The rationale is simple: understanding the demographics of those waiting for NHS care can shed light on potential disparities in access.

With the government seeking to “bring the best of the NHS to the rest”, understanding this nuanced picture of national waiting times can galvanise both national and local service leaders to look deeper into waiting lists. This can help policy makers make choices about where to channel limited resources and to design tailored, targeted interventions to close unjustified gaps in access to care for particular groups of patients.

Waits in emergency departments

Using the Emergency Care Data Set (ECDS), we looked at attendances at Type 1 and Type 2 A&E units 2 and waiting times for different people. For more information on inclusions and exclusions, see the 'About this data' section.

How long overall do people wait when attending A&E?

From April 2022 to March 2024 there were over 33 million attendances to A&E units in England. Of these attendances, 43% of people waited in A&E for more than four hours from arrival to admission, transfer or discharge, and 10% of people waited more than 12 hours.

21% of A&E attendances resulted in an admission to hospital (almost seven million). People who were admitted waited longer on average than those not admitted – an average of 10 hours and 22 minutes. Over 73% of admitted attendees waited longer than four hours, and approximately 31% waited longer than 12 hours. In comparison, people who attended A&E who were not ultimately admitted waited a considerably shorter amount of time – averaging four hours and 20 minutes. Overall, 35% of non-admitted attendees waited longer than four hours, and just 4% waited longer than 12 hours.

Patients with psycho-social or behavioural problems wait longer than other patients, whether they are eventually admitted or not

The chief complaint describes the main reason people attend A&E, as assessed by a clinician, categorised into 15 groups. Figure 1 shows the average time spent in A&E departments by complaint and admission status (including the 13% of attendances with an invalid or missing complaint code).

Amongst all attendees, patients presenting with psychosocial or behavioural problems had the longest waits. This group of patients waited an average of 11 hours and 59 minutes if they were admitted, and seven hours and 19 minutes if they were not.

Among admitted patients, obstetrics and gynaecology patients waited the shortest amount of time in A&E

Regardless of the reason that someone attends A&E, the biggest driver of their wait is whether they are ultimately admitted or not, with admitted patients waiting longer for all complaints. Although for obstetrics and gynaecology complaints, being admitted had the least impact on how long someone waits as waits for those complaints had the smallest difference between admitted and non-admitted average waits. Obstetrics and gynaecology patients waited four hours and seven minutes if admitted, and three hours and 39 minutes if not admitted.

Older people spend longer in A&E, even if they aren’t admitted

Across all age groups, attendances by 0-9 year-olds made up the highest proportion of all attendances (16%), but the probability of waiting more than four or 12 hours rises steeply with age (Figure 2). 

While longer waits in older patients can be partly explained by a higher likelihood of being admitted, our analysis shows that this pattern persists amongst admitted and non-admitted patients. We found that over one in 10 of attendees aged 80 or over who were not ultimately admitted waited longer than 12 hours in A&E.

More people from deprived areas attend A&E, but waiting times are similar across patients from all levels of deprivation

Indices of Multiple Deprivation (IMD) deciles are used to understand relative deprivation across areas in England 3 . In this section attendances are split into ten groups, where decile 1 represents patients from the most deprived areas, and decile 10 the least deprived.

Figure 3 illustrates how attendances from the most deprived areas were higher and fell steadily as deprivation decreased. However average waits varied relatively little between IMD deciles. These patterns held when considering admitted and non-admitted attendances separately.

Black, Asian and mixed ethnicity children wait longer than their White peers

ECDS collects the self-reported ethnicity of attendees on arrival. Almost 12% of attendances had missing ethnicity data reflecting a need identified in many health datasets for ethnicity coding to improve.

White patients waited longer on average – approximately five hours and 54 minutes compared to patients from Black, Asian, mixed race and other ethnic groups, who waited on average almost 60 minutes less. However, the White population in England is relatively older than minority ethnic groups, and so would be expected to have longer waits.

When we look at each age group separately (Figure 4), we see that White patients are not necessarily waiting longer across the board. Among people 60 and older, White patients still wait slightly longer, reflecting the older age distribution: even within people aged 60 and over, the White population will have more people aged 80 or over. When we look at the youngest age group (those aged 19 and under), patients of Black ethnicities waited longer – on average three hours and 41 minutes – compared to White children and young people, who waited approximately three hours and 20 minutes on average. Further analysis by ten-year age bands revealed longer waits for Black patients up until the age of 40. Amongst patients who were admitted, Black people of all ages were found to wait longer than White people.

Our analysis offers no clear explanation for these disparities – via deprivation or discharge destination. Our work also did not involve a multivariate analysis to understand the relationship between these findings and other issues such as the wide variations in patterns of illness and service use within and between ethnic groups, or the well-documented existence of structural racism in the NHS and wider society. But understanding how such factors interact with ethnicity and waiting times would be an important next step in tackling these disparities.

Waiting for planned care

Using information collected from the Referral to Treatment (RTT) Waiting Times and a large scale survey by Office of National Statistics asking people about waiting for NHS care, here we look at planned care waiting times for different services and different people.

Planned hospital care

Waiting list sizes for planned NHS hospital care and their waiting times have increased in England

In May 2024, 6.37 million people were still waiting for planned NHS hospital care. Table 1 shows the total waiting list size 4  and waiting times at three timepoints: May 2024, May 2014 and May 2021 (the first May after the estimated first and second waves of Covid-19 infections). 

Table 1: Total waiting list size for planned NHS hospital care and overall waiting times
 
Total waiting list size (mill)
Total waits within 18 weeks (mill)
Percentage of total waiting list that were waiting less than 18 weeks
Total waits that were 52 weeks or more
Percentage of total waiting list that were 52 weeks or more
Median waiting time (weeks)

May 2014

3.2

3.0

94.6%

776

0.02%

6.2

May 2021

5.3

3.6

67.4%

336,779

6.30%

10.8

May 2024

7.6

4.5

59.1%

307,500

4.00%

14.2

Source: NHS England’s Incomplete Commissioner Referral to Treatment (RTT) Waiting Times.  Note that all values take into account missing data. The median waiting time is the time in weeks where half the number of waits will be waiting longer than this.
 

The total waiting list size has more than doubled over the last decade and waiting times have increased. Despite 1.5 million more waits being within 18 weeks in May 2024 compared to May 2014, the proportion of all waits that are within that timeframe has decreased, and one in 20 of the waits people endured were over a year.

However, there are differences in waiting list sizes and waiting times depending on what people are waiting for.

Waiting list sizes and waiting times, and how these have changed over time, vary by what people are waiting for

Figure 5 shows the total waiting list size, median waiting times and the waits that are: within 18 weeks, more than 18 weeks but within 52 weeks, and 52 weeks or more in May 2024 for each specialty. 5

There is large variation in the size of the waiting lists and the waiting time between specialties. The six specialties with the largest list sizes make up 51% of the total waiting list, four of which are surgical services. 6  Specialties with larger absolute list sizes are more likely to have greater numbers of waits that are 52-plus weeks. 7

Trauma and orthopaedics, and ear, nose and throat services have the largest waiting lists

Trauma and orthopaedics services, and ear, nose and throat services, both of which are surgical, had the largest waiting lists in May 2024, making up nearly one fifth of the total waiting list size (11.0% and 8.6% respectively), and some of the worst performances 8  (Figure 5) and largest increases in waiting times (Figure 6).

Trauma and orthopaedics services are more likely to treat people requiring overnight admission and longer lengths of stay in hospital than surgical specialties where more patients requiring planned care can be treated as day cases, limiting the capacity to treat more patients. The pandemic may have negatively affected access to these types of services more, as in the first wave of the pandemic, there were greater relative drops in the number of admitted patients compared to non-admitted patients. The differing need for rehabilitation may also be a factor. 

For example, hip and knee replacement surgery, a high volume operation for which there are long waits, has been slower to return to pre-pandemic volumes compared with cataract surgery.

Figure 6 shows, for each specialty, the percentage change in their waiting list size and the change in their median wait times for May 2024 compared to May 2014 or May 2021 baselines. 9

Waiting list sizes have increased for all specialties compared to their baselines, apart from general internal medicine and mental health services. There is, however, large variation in the size of the increases, ranging from a 13% increase in waiting list size for cardiothoracic surgery services to a 263% increase for respiratory medicine services. Median waiting times increased for all specialties.

Respiratory medicine and gynaecology services have had the largest increases in waiting list sizes

Although one of the smaller waiting list sizes, respiratory medicine services saw the largest increase in total waiting list size with a 263% increase between May 2014 and May 2024, and gynaecology services had the second largest with a 223% increase between May 2014 and May 2024.

For respiratory medicine, this might be a consequence of increasing emergency care needs diverting resources away from planned care as a result of direct Covid-19 infections or exacerbations of respiratory diseases due to changes in the occurrence of respiratory viruses post pandemic, such as respiratory syncytial virus (RSV). In addition, with the emergence of long Covid, the respiratory medicine workforce are caring for a whole new group of patients suffering long-term symptoms following a Covid-19 infection.

Gynaecology services had the third largest number of waits that were 52 plus weeks, with one in 20 of its waits waiting this long in May 2024. The 2022 Elective Recovery Plan for the NHS focused on prioritisation based on clinical need and to reduce long waits over 65 weeks. This has been criticised as not effectively prioritising these services and may have resulted in waits for gynaecological services  not improving as much as orthopaedic services which had more of the very longest waiters.

What do the public tell us about waiting for planned care?

The RTT Waiting Times data discussed above tells us what planned hospital care people are waiting for but it does not tell us who is waiting for planned care.

A survey 10  in early 2024 by the Office for National Statistics (ONS) of just over 89,000 people aged 16 years old and over can help us understand this. The survey asked respondents if they are “currently waiting for a hospital appointment, test, or to start receiving medical treatment through the NHS” 11  (which will include waiting for planned hospital care but also care outside of the hospital, such as community mental health care and care from a GP practice), and if so, how long they have been waiting. 12

One fifth of the English population are estimated to be waiting for some form of NHS care

Over one in five (21%) respondents 13  said they are currently waiting for planned NHS care, which is estimated to be nearly 9.8 million people. 14  This is larger than the RTT waiting list size for planned hospital care reflecting both how people experience waiting versus how they are officially measured, such as waits for planned hospital care not captured by the RTT data set (e.g. non-consultant lead care or post treatment follow up consultations) and care outside hospital.

Although not directly comparable, how long people say they are waiting broadly reflects the RTT waiting times – an estimated 71% of respondents have been waiting for six months or less and an estimated 15% have been waiting for 12 months or longer. 

Older people and women are more likely to say they are waiting for planned NHS care, and younger people are less likely to say they have shorter waits

Despite those aged 85 and older having the largest proportion of respondents who say they are waiting for planned NHS care, the 65-74 age band had the largest number of people saying they were waiting, with over two million estimated to be waiting.

Women are more likely to say they are waiting for care than men. This is driven by women of a younger age and may partly relate to the waiting list for planned hospital-based gynaecology care.

Younger people are less likely to say they have shorter waits for planned NHS care – an estimated 60% of 16-24 year-olds say they have been waiting less than six months compared with an estimated 75% of 55-64 year-olds who say that they have been waiting this long. 15  The opposite pattern is seen for respondents waiting 12 months or longer, but there is less certainty around this due to the sample sizes at the extreme age bands. 

People from the most deprived areas are more likely to say they are waiting for care and waiting longer but there isn’t a clear deprivation gradient

Figure 10 shows how the proportion of people who report waiting for planned NHS care varies by how deprived the area they live in is. Although official data shows there are lower levels of planned care activity for people from the most deprived areas, people who live in the 20% most deprived areas are significantly more likely to say they are waiting for planned NHS care than those from other areas. And those living in the 10% least deprived areas are significantly less likely to say they are waiting than those from all other areas. This may be a reflection of the relationship between deprivation and overall health status, access to health care, how easy people find it to navigate the NHS system, and differing private health care use. 

People from the most deprived areas are also significantly less likely to say they have been waiting less than six months for their longest wait, and significantly more likely to say they have been waiting 12 months or longer compared to those from the least deprived. But there is no clear pattern in responses from people who live in areas of deprivation between these extremes.

Some minority ethnic groups are more likely to say they are waiting for planned NHS care compared to White ethnic groups

People of Black, African, Caribbean or Black British ethnicities and people of Mixed or Multiple ethnicities are significantly more likely to say they are waiting for planned NHS care than people of White ethnicities (27% and 26% respectively compared to 21%), once age differences in the populations have been accounted for. But there is no difference between people of White ethnicities and people of Asian or Asian British ethnicities or between White ethnicities and other ethnic groups. This could be partly explained by the deprivation findings and the higher proportion of people in Black and Mixed ethnic groups living in areas of high deprivation

Conclusion

The new Secretary of State for Health and Social Care has said that the NHS is broken. Our own QualityWatch analysis shows that the NHS has been struggling for over a decade, with endemic issues in access to care, targets not being met and inequalities in who gets care and how fast they get it.

Behind those headline numbers are millions of people impacted on a daily basis by delays in treatment and difficulty accessing care – the reality of which means people are waiting in pain on waiting lists, have a reduced quality of life, and are at higher risk of dying from delays in overcrowded emergency departments. Health care prioritisation is difficult, but our analysis has provided further insights behind the hospital waiting figures illustrating that not all people experience waiting for care equally:

  • What people are waiting for influences their waiting time, with on-going capacity issues both within and outside of hospitals impacting on some services more than others.
  • Older people are more likely to spend longer in A&E and are more likely to be waiting for planned NHS care, but younger people are more likely to say they have longer waits for planned care.
  • People from the most deprived areas are more likely to be waiting for care.
  • Experience of waiting varies between ethnic groups.

Prior to the election, we set out what the NHS needs from the government to improve access to care. Addressing inequalities in access needs to be prioritised, and more targeted action is needed to reduce the variation in waiting between groups and across conditions. For example, while surgical hubs show promise for speeding up care, they are currently focusing on orthopaedics, cataracts and high-volume procedures. Different solutions are likely to be needed to address areas of growing demand, such as for gynaecology services.

Local health leaders with access to more detailed data need to use it to understand who is using services and where bottlenecks exist across the system. For example, analysing the number of people coming onto the waiting list compared to number of people coming off it could indicate whether the increases are a result of changes to demand (influenced by changes in public awareness, actual prevalence and referral practices), rather than changes in ‘throughput’ (influenced by capacity and productivity).

The recent Darzi report set out the scale of the challenge facing the NHS, and validates the concerns of the public, staff and the service. This briefing has focused on A&E and waits for planned hospital care, two areas of care where the NHS is struggling, , and which are well-known by the public and politicians. The causes of these problems are multi-factorial, influenced by rising demand from a growing and sicker population; squeezed NHS budgets; hospitals lacking capacity; inadequate digital technology; and crumbling buildings.

While the government has acknowledged an urgent need to ‘fix’ the problem, solutions are complex and will not be quick to deliver. The reality is that the British public should remain braced for long waits for the care they need for some time to come.

1

See the About this data section at the end of the report for more detail on the data sets we used and how the analysis was carried out.

2

Type 1 A&E units are major A&E departments which can treat a range of problems, whereas Type 2 refers to single specialty departments (e.g. ophthalmology, dental). This analysis excludes Type 3 A&E departments which treat minor injuries and illnesses: see https://digital.nhs.uk/data-and-information/publications/statistical/hospital-accident--emergency-activity/2022-23/about-this-publication#department-…

3

The Indices of Deprivation are a unique measure of relative deprivation at a small local area level (Lower-layer Super Output Areas) across England. A decile is one of ten equal groups into which a population can be divided according to the distribution of values. See https://assets.publishing.service.gov.uk/media/5dfb3d7ce5274a3432700cf3/IoD2019_FAQ_v4.pdf

4

The number of people still waiting is lower than the total waiting list size because some people are waiting for more than one hospital appointment. The number of people still waiting has only been collected since September 2021.

5

Specialties are divisions of clinical work which may be defined by body systems (dermatology), age (paediatrics), clinical technology (nuclear medicine), clinical function (rheumatology), group of diseases (oncology) or combinations of these factors. See https://archive.datadictionary.nhs.uk/DD%20Release%20March%202023/attributes/main_specialty_code.html

6

DCB0028: Treatment Function and Main Specialty Standard – NHS England Digital see supporting document: Code List Specification v1.2 (Amd 45/2019) accessed 16 August 2024.

7

For information on how this was assessed, see the About this data section at the end of this report.

8

Performance both against percentage of waits within 18 weeks and percentage of waits 52 weeks or more.

9

Baselines were chosen as either May 2014 (a decade beforehand) or May 2021 (the earliest year they were reported on). Please see About this data for more information.

10

See the About this data section at the end of this report for further detail.

11

For the remainder of this report we refer to this as planned NHS care.

12

People who were waiting to start treatment for multiple conditions were asked to respond about their longest wait.

13

All percentages presented including the 95 per cent confidence intervals in the figures are age-standardised estimates. See the data notes for more information on the methodology.

14

All estimated numbers of people presented are based on the age-standardised estimates of the proportion of the group responding a certain way and the total weighted sample size for that particular group of people from the survey.

15

The 55-64 year-old age group is the first older age group where responses on how long people are waiting becomes significantly different from the next youngest age band when looking at those saying they have been waiting six months or less.

About this data

NHS England, Emergency Care Data Set

The Emergency Care Data Set (ECDS) is the national data set for urgent and emergency care. The data set contains details of all A&E attendances at NHS hospitals in England, including minor injury units and NHS walk-in centres, as well as 24 hour and consultant-led emergency departments. It collects information about when, how and why people attend emergency departments and what happens to them, as well as information on the patient. It is used for non-clinical purposes, such as research and planning health services.

Emergency Care Data Set data (year range 2022/23–2023/24) Copyright © (2024), NHS England. Re-used with the permission of NHS England. All rights reserved.

For more information on our information security and data use, including using patient data in research, please see our Information security and data page.

Analysis

The ECDS sample used for this analysis excluded the following attendances:

  • Planned follow ups
  • Patients dead on arrival
  • Patients admitted urgently to the mortuary or hospice
  • Patients with missing or negative wait times.

Analysis was also performed at the provider level to evaluate the data quality. The Bedfordshire Hospitals NHS Trust was found to have implausible wait times in the first year of data used and was therefore excluded from the sample for this analysis.

Attendances with a wait time of more than 72 hours were also excluded, in line with the NHS methodology for calculating performance statistics.

Derived variables

Attendances were categorised into ‘chief complaint categories’ using chief complaint SNOMED codes, and admission status was derived from attendance destination codes. Specific details of this coding can be found in sections 13.4 and 26.4 of the ECDS Enhanced Technical Output Specification (ETOS v4.0.7).

Attendee’s ethnicities were categorised into broad categories using the categorisation described by NHS England. These broad categorisations do not capture crucial differences within ethnic groups, and may poorly represent patients from certain ethnic groups, such as Chinese or Arab attendees. However, using these groups was necessary to prevent disclosure due to a smaller sample amongst minority ethnic groups.

NHS England, Referral to Treatment (RTT) Waiting Times

The Referral to Treatment (RTT) Waiting Times data monitors the number of people waiting for Consultant-led treatment, how many waits there are in total, and the length of time from a referral through to elective (planned) treatment.

Analysis

The analysis uses data on the commissioner incomplete pathways which looks at waits for patients who are still waiting to start treatment at the end of the month.

May 2024 was used as this was the most recent data at the time of analysis (date accessed 25/07/2024). May 2024 data was unrevised data at the time of access and publication, all other months have been revised.

For total waiting list size and overall waiting times in table 1 above, this used the estimated values that take missing data into account.

To assess how strongly and in what direction the specialty total waiting list sizes and number of those waits waiting 52 plus weeks relate to each other, a Pearson Correlation Coefficient was calculated as 0.8768. A value between 0 and 1 indicates when one measure changes, the other measure changes in the same direction, and the closer to one the stronger the relationship.

Where specialty names have been updated over time, the analysis reports on their 2024 specification. These are as follows.

2024

2014

Ear, Nose and Throat ServiceENT
General Internal Medicine ServiceGeneral Medicine
Respiratory Medicine ServiceThoracic Medicine
Elderly Medicine ServiceGeriatric Medicine

Changes in waiting list sizes and median waiting times by specialty compared May 2024 to a decade before in May 2014 for all specialities apart from those that are classed as ‘Other’. These compared May 2024 to May 2021 data as that was the first May that they were separated out into individual categories.

Office for National Statistics, NHS waiting times survey

The Office for National Statistics’ (ONS) NHS waiting times survey is part of the GP Access Survey from the Office for National Statistics (date accessed: 4 July 2024).

The sample consists of adults aged 16 years and over living in England, based on participants of the Winter Coronavirus (COVID-19) Infection Survey (CIS), between 16 January to 15 February 2024, who consented to completing additional voluntary questions commissioned by NHS England.

Analysis

Information on the samples for each demographic breakdown is provided in the raw data. We assessed this against population distributions, and there were smaller proportions of respondents from non-White ethnicities or lower IMD deciles than what would be expected. However, ONS provided assurances that this is typical for any voluntary social survey. Plus, the Winter COVID-19 CIS is seen as a gold standard surveillance strategy and the data is also weighted and measures of uncertainty are provided to help mitigate against issues this may cause, please see below.

All percentages presented, including the 95% confidence intervals, are the age standardised estimates provided by ONS. Age-standardised rates allow comparisons between populations that may contain proportions of different ages. They are a better measure than simply looking at crude proportions, as they take into account the population size and age structure.

All estimated numbers of people presented are based on the age-standardised estimates of the proportion of the group responding a certain way and the total weighted sample size. These are provided by ONS for that particular group of people from the survey. 'Weighted count' provides the representative count for each breakdown; this also takes into account survey design and non-response. For further information on ONS weighting, please see GP Access Survey Quality and Methodology Information.

95% confidence intervals provided by ONS are shown, which is a measure of the statistical precision of an estimate and shows the range of uncertainty around the calculated estimate. As a general rule, if the confidence interval around one figure overlaps with the interval around another, we cannot say with certainty that there is more than a chance difference between the two figures. Differences should be considered alongside confidence intervals provided. The statistical significance of differences can be determined based on non-overlapping confidence intervals.